39

1. Regular Expression Tutorial
In this tutorial, I will teach you all you need to know to be able to craft powerful time-saving regular expressions. I will start with the most basic concepts, so that you can follow this tutorial even if you know nothing at all about regular expressions yet. But I will not stop there. I will also explain how a regular expression engine works on the inside, and alert you at the consequences. This will help you to understand quickly why a particular regex does not do what you initially expected. It will save you lots of guesswork and head-scratching when you need to write more complex regexes.

What Regular Expressions Are Exactly - Terminology
Basically, a regular expression is a pattern describing a certain amount of text. Their name comes from the mathematical theory on which they are based. But we will not dig into that. Since most people including myself are lazy to type, you will usually find the name abbreviated to regex or regexp. I prefer regex, because it is easy to pronounce the plural "regexes". In this book, regular expressions are printed guillemots: «regex». They clearly separate the pattern from the surrounding text and punctuation. This first example is actually a perfectly valid regex. It is the most basic pattern, simply matching the literal text „regex”. A "match" is the piece of text, or sequence of bytes or characters that pattern was found to correspond to by the regex processing software. Matches are indicated by double quotation marks, with the left one at the base of the line. «\b[A-Z0-9._%-]+@[A-Z0-9._%-]+\.[A-Z0-9._%-]{2,4}\b» is a more complex pattern. It describes a series of letters, digits, dots, percentage signs and underscores, followed by an at sign, followed by another series of letters, digits, dots, percentage signs and underscores, finally followed by a single dot and between two and four letters. In other words: this pattern describes an email address. With the above regular expression pattern, you can search through a text file to find email addresses, or verify if a given string looks like an email address. In this tutorial, I will use the term "string" to indicate the text that I am applying the regular expression to. I will indicate strings using regular double quotes. The term “string” or “character string” is used by programmers to indicate a sequence of characters. In practice, you can use regular expressions with whatever data you can access using the application or programming language you are working with.

Different Regular Expression Engines
A regular expression “engine” is a piece of software that can process regular expressions, trying to match the pattern to the given string. Usually, the engine is part of a larger application and you do not access the engine directly. Rather, the application will invoke it for you when needed, making sure the right regular expression is applied to the right file or data. As usual in the software world, different regular expression engines are not fully compatible with each other. It is not possible to describe every kind of engine and regular expression syntax (or “flavor”) in this tutorial. I will focus on the regex flavor used by Perl 5, for the simple reason that this regex flavor is the most popular

40 one, and deservedly so. Many more recent regex engines are very similar, but not identical, to the one of Perl 5. Examples are the open source PCRE engine (used in many tools and languages like PHP), the .NET regular expression library, and the regular expression package included with version 1.4 and later of the Java JDK. I will point out to you whenever differences in regex flavors are important, and which features are specific to the Perl-derivatives mentioned above.

Give Regexes a First Try
You can easily try the following yourself in a text editor that supports regular expressions, such as EditPad Pro. If you do not have such an editor, you can download the free evaluation version of EditPad Pro to try this out. EditPad Pro's regex engine is fully functional in the demo version. As a quick test, copy and paste the text of this page into EditPad Pro. Then select Edit|Search and Replace from the menu. In the search pane that appears near the bottom, type in «regex» in the box labeled “Search Text”. Mark the “Regular expression” checkbox, unmark “All open documents” and mark “Start from beginning”. Then click the Search button and see how EditPad Pro's regex engine finds the first match. When “Start from beginning” is checked, EditPad Pro uses the entire file as the string to try to match the regex to. When the regex has been matched, EditPad Pro will automatically turn off “Start from beginning”. When you click the Search button again, the remainder of the file, after the highlighted match, is used as the string. When the regex can no longer match the remaining text, you will be notified, and “Start from beginning” is automatically turned on again. Now try to search using the regex «reg(ular expressions?|ex(p|es)?)». This regex will find all names, singular and plural, I have used on this page to say “regex”. If we only had plain text search, we would have needed 5 searches. With regexes, we need just one search. Regexes save you time when using a tool like EditPad Pro. If you are a programmer, your software will run faster since even a simple regex engine applying the above regex once will outperform a state of the art plain text search algorithm searching through the data five times. Regular expressions also reduce development time. With a regex engine, it takes only one line (e.g. in Perl, PHP, Java or .NET) or a couple of lines (e.g. in C using PCRE) of code to, say, check if the user's input looks like a valid email address.

41

2. Literal Characters
The most basic regular expression consists of a single literal character, e.g.: «a». It will match the first occurrence of that character in the string. If the string is “Jack is a boy”, it will match the „a” after the “J”. The fact that this “a” is in the middle of the word does not matter to the regex engine. If it matters to you, you will need to tell that to the regex engine by using word boundaries. We will get to that later. This regex can match the second „a” too. It will only do so when you tell the regex engine to start searching through the string after the first match. In a text editor, you can do so by using its “Find Next” or “Search Forward” function. In a programming language, there is usually a separate function that you can call to continue searching through the string after the previous match. Similarly, the regex «cat» will match „cat” in “About cats and dogs”. This regular expression consists of a series of three literal characters. This is like saying to the regex engine: find a «c», immediately followed by an «a», immediately followed by a «t». Note that regex engines are case sensitive by default. «cat» does not match “Cat”, unless you tell the regex engine to ignore differences in case.

Special Characters
Because we want to do more than simply search for literal pieces of text, we need to reserve certain characters for special use. In the regex flavors discussed in this tutorial, there are 11 characters with special meanings: the opening square bracket «[», the backslash «\», the caret «^», the dollar sign «$», the period or dot «.», the vertical bar or pipe symbol «|», the question mark «?», the asterisk or star «*», the plus sign «+», the opening round bracket «(» and the closing round bracket «)». These special characters are often called “metacharacters”. If you want to use any of these characters as a literal in a regex, you need to escape them with a backslash. If you want to match „1+1=2”, the correct regex is «1\+1=2». Otherwise, the plus sign will have a special meaning. Note that «1+1=2», with the backslash omitted, is a valid regex. So you will not get an error message. But it will not match “1+1=2”. It would match „111=2” in “123+111=234”, due to the special meaning of the plus character. If you forget to escape a special character where its use is not allowed, such as in «+1», then you will get an error message. All other characters should not be escaped with a backslash. That is because the backslash is also a special character. The backslash in combination with a literal character can create a regex token with a special meaning. E.g. «\d» will match a single digit from 0 to 9.

Special Characters and Programming Languages
If you are a programmer, you may be surprised that characters like the single quote and double quote are not special characters. That is correct. When using a regular expression or grep tool like PowerGREP or the

42 search function of a text editor like EditPad Pro, you should not escape or repeat the quote characters like you do in a programming language. In your source code, you have to keep in mind which characters get special treatment inside strings by your programming language. That is because those characters will be processed by the compiler, before the regex library sees the string. So the regex «1\+1=2» must be written as "1\\+1=2" in C++ code. The C++ compiler will turn the escaped backslash in the source code into a single backslash in the string that is passed on to the regex library. To match „c:\temp”, you need to use the regex «c:\\temp». As a string in C++ source code, this regex becomes "c:\\\\temp". Four backslashes to match a single one indeed. See the tools and languages section in this book for more information on how to use regular expressions in various programming languages.

Non-Printable Characters
You can use special character sequences to put non-printable characters in your regular expression. «\t» will match a tab character (ASCII 0x09), «\r» a carriage return (0x0D) and «\n» a line feed (0x0A). Remember that Windows text files use “\r\n” to terminate lines, while UNIX text files use “\n”. You can include any character in your regular expression if you know its hexadecimal ASCII or ANSI code for the character set that you are working with. In the Latin-1 character set, the copyright symbol is character 0xA9. So to search for the copyright symbol, you can use «\xA9». Another way to search for a tab is to use «\x09». Note that the leading zero is required.

can only be implemented in regex-directed engines. Again. will the engine continue with the second character in the text. At that point. There are two kinds of regular expression engines: text-directed engines. No surprise that this kind of engine is more popular. then it is text-directed.”. The entire regular expression could be matched starting at character 15. the engine will start at the first character of the string. and regex-directed engines. The engine is "eager" to report a match. «t» fails to match “p”. It will try all possible permutations of the regular expression at the first character. there are a few versions of these tools that use a regex-directed engine. If the resulting match is only „regex”. The reason behind this is that the regex-directed engine is “eager”. after introducing a new regex token. If backreferences and/or lazy quantifiers are available. I will explain step by step how the regex engine actually processes that token. When applying a regex to a string. lex. «c» fails to match here and the engine carries on. „a”. the engine will try to match the first token in the regex «c» to the first character in the match “H”. «c» matches „c”. If the result is „regex not”. This is because certain very useful features. MySQL and Procmail. This succeeds too. the engine knows the regex cannot be matched starting at the 4th character in the match. All the regex flavors treated in this tutorial are based on regex-directed engines. So it will continue with the 5th: “a”. The engine will then try to match the second token «a» to the 5th character. such as lazy quantifiers and backreferences. But then. The engine then proceeds to attempt to match the remainder of the regex at character 15 and finds that «a» matches „a” and «t» matches „t”. It will help you understand quickly why a particular regex does not do what you initially expected.43 3. because it merely consists of a sequence of literal characters. Jeffrey Friedl calls them DFA and NFA engines. flex. Only if all possibilities have been tried and found to fail. it will try all possible permutations of the regex. as does matching the «c» with the space. You can do the test by applying the regex «regex|regex not» to the string “regex not”. First Look at How a Regex Engine Works Internally Knowing how the regex engineworks will enable you to craft better regexes more easily. egrep. This inside look may seem a bit long-winded at certain times. the engine is regex-directed. «c» again matches „c”. There are no other possible permutations of this regex. . Notable tools that use text-directed engines are awk. Arriving at the 4th character in the match. But understanding how the regex engine works will enable you to use its full power and help you avoid common mistakes. you can be certain the engine is regex-directed. even if a “better” match could be found later. It will therefore report the first three letters of catfish as a valid match. This will save you lots of guesswork and head-scratching when you need to write more complex regexes. The Regex-Directed Engine Always Returns the Leftmost Match This is a very important point to understand: a regex-directed engine will always return the leftmost match. Again. At the 15th character in the match. in exactly the same order. In this tutorial. The result is that the regex-directed engine will return the leftmost match. This fails. The engine never proceeds beyond this point to see if there are any “better” matches. When applying «cat» to “He captured a catfish for his cat. respectively. The first match is considered good enough. So the regex engine tries to match the «c» with the “e”. For awk and egrep. This fails too. You can easily find out whether the regex flavor you intend to use has a text-directed or regex-directed engine.

44 In this first example of the engine's internals. In following examples. Some of the results may be surprising. our regex engine simply appears to work like a regular text search routine. But they are always logical and predetermined. once you know how the engine works. . A text-directed engine would have returned the same result too. it is important that you can follow the steps the engine takes in your mind. the way the engine works will have a profound impact on the matches it will find. However.

You can use more than one range. Character Classes or Character Sets With a "character class". Unlike the dot. and only the q. Your regex will work fine if you escape the regular metacharacters inside a character class. «[0-9]» matches a single digit between 0 and 9. «[0-9a-fA-F]» matches a single hexadecimal digit. in both strings. the backslash (\). Find an identifier in a programming language with «[A-Za-z_][A-Za-z_0-9]*». «q[^u]» does not mean: “a q not followed by a u”. You can combine ranges and single characters. «gr[ae]y» will not match “graay”. Indeed: the space will be part of the overall match. the caret (^) and the hyphen (-). and do not need to be escaped by a backslash. The order of the characters inside a character class does not matter. It will not match the q in the string “Iraq”. «[0-9a-fxA-FX]» matches a hexadecimal digit or the letter X. It will match the q and the space after the q in “Iraq is a country”. Useful Applications Find a word. use «[+*]». Metacharacters Inside Character Classes Note that the only special characters or metacharacters inside a character class are the closing bracket (]). The result is that the character class will match any character that is not in the character class. You could use this in «gr[ae]y» to match either „gray” or „grey”. Find a C-style hexadecimal number with «0[xX][A-Fa-f0-9]+». such as «sep[ae]r[ae]te» or «li[cs]en[cs]e».45 4. Negated Character Classes Typing a caret after the opening square bracket will negate the character class. It means: “a q followed by a character that is not a u”. The results are identical. you need to use negative lookahead: «q(?!u)». even if it is misspelled. Again. It is important to remember that a negated character class still must match a character. A character class matches only a single character. The usual metacharacters are normal characters inside a character class. “graey” or any such thing. To search for a star or plus. But we will get to that later. You can use a hyphen inside a character class to specify a range of characters. Very useful if you do not know whether the document you are searching through is written in American or British English. . If you want to match an a or an e. but doing so significantly reduces readability. also called “character set”. the order of the characters and the ranges does not matter. because it is the “character that is not a u” that is matched by the negated character class in the above regexp. you can tell the regex engine to match only one out of several characters. use «[ae]». case insensitively. negated character classes also match (invisible) line break characters. Simply place the characters you want to match between square brackets. If you want the regex to match the q.

Some flavors include additional. The hyphen can be included right after the opening bracket. you can see the characters matched by «\w» in PowerGREP when using the Western script. which characters this actually includes. word characters from other languages may also match. rarely used non-printable characters such as vertical tab and form feed. «[\s\d]» matches a single character that is either whitespace or a digit. «[]x]» matches a closing bracket or an x. «[^]x]» matches any character that is not a closing bracket. it includes «[ \t]». the caret (^) and the hyphen (-) can be included by escaping them with a backslash. Exactly which characters it matches differs between regex flavors. Negated Shorthand Character Classes The above three shorthands also have negated versions. In some flavors. That is: «\s» will match a space or a tab. Both «[-x]» and «[x-]» match an x or a hyphen. «[x^]» matches an x or a caret. «\D» is the same as «[^\d]». etc. and is equivalent to «[0-9a-fA-F]». you have to escape it with another backslash. while the latter matches „1” (one). «[\da-fA-F]» matches a hexadecimal digit. In the screen shot. characters with diacritics used in languages such as French and Spanish will be included. Shorthand Character Classes Since certain character classes are used often. or right before the closing bracket. Shorthand character classes can be used both inside and outside the square brackets. «\s\d» matches a whitespace character followed by a digit. «\W» is short for «[^\w]» and «\S» is the equivalent of «[^\s]». for example. it will include «[A-Za-z]». If you are using the Western script. To include a caret. You can put the closing bracket right after the opening bracket. The closing bracket (]). I recommend the latter method. In EditPad Pro. When applied to “1 + 2 = 3”. the underscore and digits are also included. «[\\x]» matches a backslash or an x. Again. In all flavors. or the negating caret. since it improves readability. the former regex will match „ 2” (space two). it also includes a carriage return or a line feed as in «[ \t\r\n]». In all flavors discussed in this tutorial. «\s» stands for “whitespace character”. Russian characters will be included. In most flavors. In most. a series of shorthand character classes are available. depends on the regex flavor. the actual character range depends on the script you have chosen in Options|Font. If you are using the Cyrillic script. place it anywhere except right after the opening bracket. or by placing them in a position where they do not take on their special meaning.46 To include a backslash as a character without any special meaning inside a character class. . «\d» is short for «[0-9]». The best way to find out is to do a couple of tests with the regex flavor you are using. or right after the negating caret. «\w» stands for “word character”.

If you want to repeat the matched character. So it will match „x”. rather than the class. It will return „grey” as the match result. but not “8”. Let us take a look at that first. The latter will match any character that is not a digit or whitespace. The last regex token is «y». and continue with the next character in the string. «[\D\S]» is not the same as «[^\d\s]». The next token in the regex is the literal «r». Repeating Character Classes If you repeat a character class by using the «?». I did not yet explain how character classes work inside the regex engine. «gr[ae]y» will match „grey” in “Is his hair grey or gray?”. Because a digit is not whitespace. When the engine arrives at the 13th character. Below. «([09])\1+» will match „222” but not “837”. The former. digit. The regex «[0-9]+» can match „837” as well as „222”. The engine will fail to match «g» at every step. „g” is matched. it must continue trying to match all the other permutations of the regex pattern before deciding that the regex cannot be matched with the text starting at character 13. The character class gives the engine two options: match «a» or match «e». it will match „3333” in the middle of this string. because that is the leftmost match. «*» or «+» operators. the leftmost match was returned. and fail. which can be matched with the following character as well. So the third token. you will need to use backreferences. Looking Inside The Regex Engine As I already said: the order of the characters inside a character class does not matter. you need to use lookahead and lookbehind. If you do not want that. and whitespace is not a digit. We already saw how the engine applies a regex consisting only of literal characters. will match any character that is either not a digit. and find that «e» matches „e”. because another equally valid match was found to the left of it. «[\D\S]» will match any character. Again. and „gray” could have been matched in the string. «[ae]» is attempted at the next character in the text (“e”). That is: «gr[ae]y» can match both „gray” and „grey”.47 Be careful when using the negated shorthands inside square brackets. But the engine simply did not get that far. whitespace or otherwise. The engine has found a complete match with the text starting at character 13. which matches the next character in the text. So it will continue with the other option. The engine will then try to match the remainder of the regex with the text. I will explain how it applies a regex that has more than one permutation. and look no further. But because we are using a regex-directed engine. even though we put the «a» first in the character class. and not just the character that it matched. you will repeat the entire character class. . When applied to the string “833337”. however. But I digress. or is not whitespace. It will first attempt to match «a». Nothing noteworthy happens for the first twelve characters in the string.

such as in Regex. so the dot could never match them. You can activate single-line mode by adding an s after the regex code. it is also the most commonly misused metacharacter. The problem is that the regex will also match in cases where it should not match.NET framework. Multi-line mode only affects anchors. the mode where the dot also matches newlines is called "single-line mode".. If you are new to regular expressions. Remember that the dot is not a metacharacter inside a character class. the first dot matched „5”. The only exception are newlinecharacters. you activate this mode by specifying RegexOptions. So by default. and single-line mode only affects the dot. you simply tick the checkbox labeled “dot matches newline”./. RegexOptions. They would read a file line by line. and apply the regular expression separately to each line. Use The Dot Sparingly The dot is a very powerful regex metacharacter. like this: m/^regex$/s. Obviously not what we intended. In Perl.\d\d». “regex”. Modern tools and languages can apply regular expressions to very large strings or even entire files. . The dot matches a single character. So if you expose this option to your users.]\d\d[. In this match.\d\d. The first tools that used regular expressions were line-based. without caring what that character is. EditPad Pro and PowerGREP. and everything will match just fine when you test the regex on valid data. The Dot Matches (Almost) Any Character In regular expressions. Other languages and regex libraries have adopted Perl's terminology. In all programming languages and regex libraries I know. the string could never contain newlines. In all regex flavors discussed in this tutorial.Match(“string”.Singleline). dot and forward slash as date separators. EditPad Pro or PowerGREP. the dot or period is one of the most commonly used metacharacters. Seems fine at first. I will illustrate this with a simple example.Singleline. and the second matched „7”. It allows you to be lazy.48 5. but we want to leave the user the choice of date separators. because it is easy to mix up this term with “multi-line mode”. It will match a date like „02/12/03” just fine. some of these cases may not be so obvious at first. All regex flavors discussed here have an option to make the dot match all characters.]\d\d» is a better solution. «\d\d[. This exception exists mostly because of historic reasons. This is a bit unfortunate. This regex allows a dash. the dot is short for the negated character class «[^\n]» (UNIX regex flavors) or «[^\r\n]» (Windows regex flavors). Put in a dot. Trouble is: „02512703” is also considered a valid date by this regular expression. In RegexBuddy. The effect is that with these tools. When using the regex classes of the . the dot will not match a newline character by default. space./. Let's say we want to match a date in mm/dd/yy format. so we do not need to escape it with a backslash. please give it a clearer label like was done in RegexBuddy. activating single-line mode has no effect other than making the dot match newlines. The quick solution is «\d\d. Unfortunately. including newlines.

The reason for this is that the star is greedy. Suppose you want to match a double-quoted string. we improved our regex by replacing the dot with a character class. It matches „99/99/99” as a valid date. but the warning is important enough to mention it here as well. and the star allows the dot to be repeated any number of times. We can have any number of any character between the double quotes. We want any number of characters that are not double quotes or newlines between the quotes. I will illustrate with an example. If you are validating user input. «[0-1]\d[. In the date-matching example. so «".49 This regex is still far from perfect.]\d\d» is a step ahead. Please respond. If you test this regex on “Put a “string” between double quotes”. . including zero. The dot matches any character. our last attempt is probably more than sufficient to parse the data without errors. Use Negated Character Sets Instead of the Dot I will explain this in depth when I present you the repeat operators star and plus. So the proper regex is «"[^"\r\n]*"». Here. How perfect you want your regex to be depends on what you want to do with it. we will do the same. Sounds easy. You can find a better regex to match dates in the example section.*"» seems to do the trick just fine. it has to be perfect. we have a problem with “string one” and “string two”. it will match „“string”” just fine. Now go ahead and test it on “Houston. though it will still match „19/39/99”.” Ouch. If you are parsing data files from a known source that generates its files in the same way every time. Our original definition of a double-quoted string was faulty./. The regex matches „“string one” and “string two””.][0-3]\d[/. Definitely not what we intended. We do not want any number of any character between the quotes.

Multiline. RegexOptions. Anchors are a different breed. the anchors match before and after newlines when you specify RegexOptions. «$» matches right after the last character in the string. like “first line\nsecond line” (where \n indicates a line break). the caret and dollar always match at the start and end of each line. you do this by adding an m after the regex code.. If you use the code if ($input =~ m/\d+/) in a Perl script to see if the user entered an integer number. In Perl. It is traditionally called "multi-line mode". putting one in a regex will cause the regex engine to try to match a single character. they match a position before. it is good practice to trim leading and trailing whitespace. the line break will also be stored in the variable. «$» will still match at the end of the string (after the last “e”). it is often desirable to work with lines. such as in Regex. See below for the inside view of the regex engine. Start of String and End of String Anchors Thus far. Handy use of alternation and /g allows us to do this in a single line of code. Likewise. you could use $input =~ s/^\s+|\s+$//g.Multiline). . rather than short strings.NET. and “end of string” must be matched right after it. Useful Applications When using regular expressions in a programming language to validate user input. and also before every line break (between “e” and “\n”). while «a$» does not match at all. using anchors is very important. This makes sense because those applications are designed to work with entire files. Similarly. The caret «^» matches the position before the first character in the string. Because “start of string” must be matched before the match of «\d+». They can be used to “anchor” the regex match at a certain position. you have to explicitly activate this extended functionality. because the «b» cannot be matched right after the start of the string. matched by «^». In every programming language and regex library I know. In . rather than the entire string. it will accept the input even if the user entered “qsdf4ghjk”. all the regex engines discussed in this tutorial have the option to expand the meaning of both anchors.Match(“string”. «^» can then match at the start of the string (before the “f” in the above string).50 6. Applying «^a» to “abc” matches „a”. Using ^ and $ as Start of Line and End of Line Anchors If you have a string consisting of multiple lines. In text editors like EditPad Pro or GNU Emacs. like this: m/^regex$/m. «^b» will not match “abc” at all. and regex tools like PowerGREP. It is easy for the user to accidentally type in a space. as well as after each line break (between “\n” and “s”). the entire string must consist of digits for «^\d+$» to be able to match. Instead. «^\s+» matches leading whitespace and «\s+$» matches trailing whitespace. because «\d+» matches the 4. They do not match any character at all. When Perl reads from a line from a text file. I have explained literal characters and character classes. In Perl. So before validating input. Therefore. «c$» matches „c” in “abc”. The correct regex to use is «^\d+$». after or between characters. “regex”. In both cases.

when reading a line from a file. and is copied by many regex flavors. use «\z» (lower case z instead of upper case Z). Looking Inside the Regex Engine Let's see what happens when we try to match «^4$» to “749\n486\n4” (where \n represents a newline character) in multi-line mode. It remains at “7”. which is not matched by the character class. «\Z» only ever matches at the end of the string. including Java.NET.51 Permanent Start of String and End of String Anchors «\A» only ever matches at the start of the string. even when you turn on “multiline mode”. then «\Z» and «$» will match at the position before that line break. We are using multi-line mode. In VB. which does not match “7”. rather than matching a character. it is common to prepend a “greater than” symbol and a space to each line of the quoted message. the resulting string will end with a line break. The first token in the regular expression is «^». «4» is a literal character. Since the match does not include any characters. both «^[a-z]+$» and «\A[a-z]+\Z» will match „joe”. Likewise. for example. «\A[a-z]+\z» does not match “joe\n”. would cause the script to accept an empty string as a valid input. In Perl. just like we want it. where the caret and dollar always match at the start and end of lines. RegexOptions. The engine then advances to the next regex token: «4». «\z» matches after the line break. These two tokens never match at line breaks. However. This means that when a regex only consists of one or more anchors. When applied to this string. "^". so the regex «^» matches at the start of the quoted message. . there is one exception. See below. In EditPad Pro and PowerGREP. «\A» and «\Z» only match at the start and the end of the entire file. As usual. Strings Ending with a Line Break Even though «\Z» and «$» only match at the end of the string (when the option for the caret and dollar to match at embedded line breaks is off). but rather with the position before the character that the regex engine has reached so far. we can easily do this with Dim Quoted as String = Regex.Multiline). "> ". This is true in all regex flavors discussed in this tutorial. this can be very useful or undesirable. Zero-Length Matches We saw that the anchors match at a position. the regex engine does not advance to the next character in the string. If the string ends with a line break. Since this token is a zero-width token. rather than at the very end of the string. the regex engine starts at the first character: “7”.Replace(Original. and after each newline.NET and PCRE. it can result in a zero-length match. the match does include a starting position. Using «^\d*$» to test if the user entered a number (notice the use of the star instead of the plus). nothing is deleted. matching only a position can be very useful. In email. If you only want a match at the absolute very end of the string. and the replacement string is inserted there. and insert the replacement string (greater than symbol and a space). The Regex. There are no other permutations of the . the engine does not try to match it with the character. «^» indeed matches the position before “7”. Since the previous token was zero-width. Reading a line from a file with the text “joe” results in the string “joe\n”. This “enhancement” was introduced by Perl. Depending on the situation. However.Replace method will remove the regex match from the string.

That fails. We already saw that those match. If you would query the engine for the character position. Let's see why. but does not advance the character position in the string. and the engine reports success. in this case. the regex engine arrives at the second “4” in the string. It does not matter that this “character” is the void after the string. So the engine arrives at «$». «4» matches „4”. the position before “\n” is preceded by a character. the engine must try to match the first token again. because MatchPosition can point to the void after the string. Again. «4». Finally. at “\n”. and that character is not a newline. it would return the length of the string if string indices are zero-based. so it will try to match the position before the current character. What you have to watch out for is that String[Regex. The engine will try to match «\d» with the void after the string. It matches the position before the void after the string. In fact. where the caret does not match. This position is preceded by a character. Another Inside Look Earlier I mentioned that «^\d*$» would successfully match an empty string. Since «$» was the last token in the regex. Not even a negated character class. As we will see later. or the void after the string. and the current character is advanced to the very last position in the string: the void after the string. If you would query the engine for the length of the match. the engine successfully matches «4» with „4”. The next attempt. but the star turns the failure of the «\d» into a zero-width success. Then. This time. “8”. It must be either a newline. The engine will proceed with the next regex token.MatchPosition] may cause an access violation or segmentation fault. we are trying to match a dollar sign. This can also happen with «^» and «^$» if the last character in the string is a newline. because it is preceded by the void before the string. The «^» can match at the position before the “4”. With success. the regex engine advances to the next regex token. and the engine advances both the regex token and the string character. and fails again. It is zero-width. the regex engine tries to match the first token at the third “4” in the string. Previously. «^» cannot match at the position before the 4. Now the engine attempts to match «$» at the position before (indeed: before) the “8”. so the engine continues at the next character. The engine continues at “9”. Yet again. the engine has found a successful match: the last „4” in the string. “9”. Again. it was successfully matched at the second “4”. the dollar matches successfully. or the length+1 if string indices are one-based in your programming language. The current regex token is advanced to «$». because this position is preceded by a character. . for «$» to match the position before the current character. also fails. the dollar will check the current character. After that. and the void after the string. There is only one “character” position in an empty string: the void after the string. The next token is «\d*». and that character is not a newline. The first token in the regex is «^». one of the star's effects is that it makes the «\d». it would return zero. and the mighty dollar is a strange beast. so the engine starts again with the first regex token. Same at the six and the newline. No regex token that needs a character to match can match here. without advancing the position in the string. at the next character: “4”. the entire regex has matched the empty string. At this point. and that character is not a newline.52 regex. because it is preceded by a newline character. However. Caution for Programmers A regular expression such as «$» all by itself can indeed match after the string. optional. The dollar cannot match here. Since that is the case after the example.

All non-word characters are always matched by «\W». if the last character is a word character. After the last character in the string. «\b» matches here because the space is not a word character. the engine continues with the «i» which does not match with the space. This regex will not match “44 sheets of a4”. The engine starts with the first token «\b» at the first character “T”. It cannot match between the “h” and the “i” either. This is because any position between characters can never be both at the start and at the end of a word. and the preceding character is. . This match is zero-length. Between a non-word character and a word character following right after the non-word character. the position before the character is inspected. It matches at a position that is called a “word boundary”. Looking Inside the Regex Engine Let's see what happens when we apply the regex «\bis\b» to the string “This island is beautiful”. so the engine retries the first token at the next character position. and neither between the “i” and the “s”. «\b» cannot match at the position between the “T” and the “h”.53 7. «\B» matches at any position between two word characters as well as at any position between two non-word characters. All characters that are not “word characters” are “non-word characters”. there is only one metacharacter that matches both before a word and after a word. «\b» matches here. Again. So «\b4\b» can be used to match a 4 that is not part of a larger number. «\B» matches at every position where «\b» does not. Word Boundaries The metacharacter «\b» is an anchor like the caret and the dollar sign. Note that «\w» usually also matches digits. «i» does not match “T”. but all word characters are always matched by the short-hand character class «\w». because the previous regex token was zero-width. Between a word character and a non-word character following right after the word character. In Perl and the other regex flavors discussed in this tutorial. because the T is a word character and the character before it is the void before the start of the string. A “word character” is a character that can be used to form words. Using only one operator makes things easier for you. Simply put: «\b» allows you to perform a “whole words only” search using a regular expression in the form of «\bword\b». if the first character is a word character. The engine does not advance to the next character in the string. Since this token is zero-length. There are four different positions that qualify as word boundaries: • • • • Before the first character in the string. Effectively. Negated Word Boundary «\B» is the negated version of «\b». The next character in the string is a space. The exact list of characters is different for each regex flavor. So saying "«\b» matches before and after an alphanumeric sequence“ is more exact than saying ”before and after a word". The engine continues with the next token: the literal «i».

«\b». the regex engine finds that «i» matches „i” and «s» matches „s”. The engine reverts to the start of the regex and advances one character to the “s” in “island”.54 Advancing a character and restarting with the first regex token. skipping the two earlier occurrences of the characters i and s. the «\b» fails to match and continues to do so until the second space is reached. If we had used the regular expression «is». Again. It matches there. Continuing. The engine has successfully matched the word „is” in our string. Now. also matches at the position before the second space in the string because the space is not a word character. But «\b» matches at the position before the third “i” in the string. but matching the «i» fails. and finds that «i» matches „i” and «s» matches «s». the engine tries to match the second «\b» at the position before the “l”. The engine continues. The last token in the regex. it would have matched the „is” in “This”. and the character before it is. . «\b» matches between the space and the second “i” in the string. This fails because this position is between two word characters.

or. as well as the next token in the regex. and then another word boundary. The match succeeds. Alternation is similar. we would need to use «\b(cat|dog)\b». If we want to improve the first example to match whole words only. Alternation with The Vertical Bar or Pipe Symbol I already explained how you can use character classes to match a single character out of several possible characters. “S”. or everything to the right of the vertical bar. we can optimize this further to «\b(Get|Set)(Value)?\b». The best option is probably to express the fact that we only want to match complete words. so the entire regex has successfully matched „Set” in “SetValue”. However. The regex engine starts at the first token in the regex. The next token in the regex is the «e» after the «S» that just successfully matched. «t» matches „t”. Suppose you want to use a regex to match a list of function names in a programming language: Get. and that the entire regex has not failed yet. If you want more options. Let's see how this works out when the string is “SetValue”. The match fails again. the regex engine would have searched for “a word boundary followed by cat”. then either “cat” or “dog”. it considers the entire alternation to have been successfully matched as soon as one of the options has. and the engine continues with the next character in the string. The obvious solution is «Get|GetValue|Set|SetValue». If you want to search for the literal text «cat» or «dog». If you want to limit the reach of the alternation. the order of the alternatives matters. «SetValue» will be attempted before «Set». and the engine will match the entire string. simply expand the list: «cat|dog|mouse|fish». there are no other tokens in the regex outside the alternation. At this point. So the solution is «\b(Get|GetValue| Set|SetValue)\b» or «\b(Get(Value)?|Set(Value)?)\b». So it knows that this regular expression uses alternation. Contrary to what we intended. Set or SetValue. you will need to use round brackets for grouping. «e» matches „e”. being the second «G» in the regex. "dog followed by a word boundary. the regex did not match the entire string. The next token. It will stop searching as soon as it finds a valid match. So it continues with the second option. There are several solutions. If we had omitted the round brackets. The next token is the first «S» in the regex. «G». One option is to take into account that the regex engine is eager. The alternation operator has the lowest precedence of all regex operators. and at the first character in the string.55 8. The consequence is that in certain situations. We could also combine the four options into two and use the question mark to make part of them optional: «Get(Value)?|Set(Value)?». . Because the question mark is greedy. the regex engine studied the entire regular expression before starting. This tells the regex engine to find a word boundary. That is. and change the order of the options. the third option in the alternation has been successfully matched. We do not want to match Set or SetValue if the string is “SetValueFunction”. «SetValue» will be attempted before «Set». In this example. The match fails. You can use alternation to match a single regular expression out of several possible regular expressions. Remember That The Regex Engine Is Eager I already explained that the regex engine is eager. it tells the regex engine to match either everything to the left of the vertical bar. If we use «GetValue|Get|SetValue|Set». Because the regex engine is eager. separate both options with a vertical bar or pipe symbol: «cat|dog». Since all options have the same end. GetValue.

This fails.: «colou?r» matches both „colour” and „color”. I will say a lot more about greediness when discussing the other repetition operators. „February 23”. Again: no problem.g. This fails. The effect is that if you apply the regex «Feb 23(rd)?» to the string “Today is Feb 23rd. will the engine try ignoring the part the question mark applies to. After a series of failures. and placing the question mark after the closing bracket. the question mark tells the regex engine that failing to match «u» is acceptable. and finds that «o» matches „o”. You can make the question mark lazy (i. The engine will always try to match that part. the engine will skip ahead to the next regex token: «r». 2003”. . „Feb 23rd” and „Feb 23”. The engine continues. However. «Feb(ruary)? 23(rd)?» matches „February 23rd”. or do not try to match it. the match will always be „Feb 23rd” and not „Feb 23”. Now. Therefore. E.56 9. «l» matches „l” and another «o» matches „o”. I have introduced the first metacharacter that is greedy. «c» will match with the „c” in “color”. The first token in the regex is the literal «c». the engine starts again trying to match «c» to the first o in “colonel”. You can write a regular expression that matches many alternatives by including more than one question mark. the engine can only conclude that the entire regular expression cannot be matched starting at the „c” in “colonel”. Only if this causes the entire regular expression to fail. Looking Inside The Regex Engine Let's apply the regular expression «colou?r» to the string “The colonel likes the color green”. «l» and «o» match the following characters. and «o».: «Nov(ember)?» will match „Nov” and „November”. E.e. Optional Items The question mark makes the preceding token in the regular expression optional. Important Regex Concept: Greediness With the question mark. Therefore. The question mark gives the regex engine two choices: try to match the part the question mark applies to. The question mark allows the engine to continue with «r». Now the engine checks whether «u» matches “r”.g. The first position where it matches successfully is the „c” in “colonel”. You can make several tokens optional by grouping them together using round brackets. This matches „r” and the engine reports that the regex successfully matched „color” in our string. turn off the greediness) by putting a second question mark after the first. But this fails to match “n” as well. Then the engine checks whether «u» matches “n”.

I did not. You might expect the regex to match „<EM>” and when continuing after that match. Only if that causes the entire regex to fail. But this regex may be sufficient if you know the string you are searching through does not contain any such invalid tags. Like the plus. If it sits between sharp brackets. Limiting Repetition Modern regex flavors. The plus tells the engine to attempt to match the preceding token once or more. «\b[1-9][09]{2. it will go back to the plus. Repetition with Star and Plus I already introduced one repetition operator or quantifier: the question mark. . it's OK if the second character class matches nothing. Obviously not what we wanted. The regex will match „<EM>first</EM>”. It tells the engine to attempt match the preceding token zero times or once. You could use «\b[1-9][0-9]{3}\b» to match a number between 1000 and 9999. matching „T”. If the comma is present but max is omitted. That is. The first character class matches a letter. Most people new to regular expressions will attempt to use «<. The sharp brackets are literals. in effect making it optional. After that. So our regex will match a tag like „<B>”. the star and the repetition using curly braces are greedy. They will be surprised when they test it on a string like “This is a <EM>first</EM> test”. Omitting both the comma and max tells the engine to repeat the token exactly min times.}» is the same as «*». because this regex would match „<1>”. The syntax is {min. The asterisk or star tells the engine to attempt to match the preceding token zero or more times. which is not a valid HTML tag. When matching „<HTML>”. will the regex engine backtrack. „</EM>”. The reason is that the plus is greedy. So «{0. Let's take a look inside the regex engine to see in detail how this works and why this causes our regex to fail. The second character class matches a letter or digit. „M” and „L” with each step. like those discussed in this tutorial. and «{1.4}\b» matches a number between 100 and 99999. Because we used the star. «<[A-Za-z][A-Za-z0-9]*>» matches an HTML tag without any attributes. I will present you with two possible solutions. it is an HTML tag. The star will cause the second character class to be repeated three times.57 10. You know that the input will be a valid HTML file. That is. and max is an integer equal to or greater than min indicating the maximum number of matches. have an additional repetition operator that allows you to specify how many times a token can be repeated.}» is the same as «+». I could also have used «<[A-Za-z0-9]+>». where min is a positive integer number indicating the minimum number of matches. the maximum number of matches is infinite. so the regular expression does not need to exclude any invalid use of sharp brackets. and proceed with the remainder of the regex. But it does not. make it give up the last iteration. The star repeats the second character class. Notice the use of the word boundaries.max}.+>». the first character class will match „H”. Watch Out for The Greediness! Suppose you want to use a regex to match an HTML tag. the plus causes the regex engine to repeat the preceding token as often as possible.

+» is reduced to „EM>first</EM> tes”. «>» cannot match here. The engine reports that „<EM>first</EM>” has been successfully matched. It will reduce the repetition of the plus by one. there is a better option than making the plus lazy. The dot is repeated by the plus. The plus is greedy. and the engine tries again to continue with «>». So far. and the engine continues with «>» and “M”. The dot will match all remaining characters in the string. so the regex continues to try to match the dot with the next character. and the engine continues repeating the dot. and the dot is repeated once more. An Alternative to Laziness In this case. Therefore. As we already know. this time repeated by a lazy plus. This tells the regex engine to repeat the dot as few times as possible. So the engine continues backtracking until the match of «. It will not continue backtracking further to see if there is another possible match. the first place where it will match is the first „<” in the string. the engine has to backtrack for each character in the HTML tag that it is trying to match. It will report the first valid match it finds. «<» matches the first „<” in the string. This fails. The total match so far is reduced to „<EM>first</EM> te”.+» has matched „<EM>first</EM> test” and the engine has arrived at the end of the string. The next character is the “>”. You can do the same with the star.+» is reduced to „EM>first</EM”. That's more like it. . We can use a greedy plus and a negated character class: «<[^>]+>». The next token in the regex is still «>». and then continue trying the remainder of the regex. But this time. The reason why this is better is because of the backtracking. But now the next character in the string is the last “t”.) Rather than admitting failure. The engine remembers that the plus has repeated the dot more often than is required. these cannot match. So the match of «. Let's have another look inside the regex engine.+?>». The dot fails when the engine has reached the void after the end of the string. „M” is matched. the engine will repeat the dot as many times as it can. which matches any character except newlines. This is a literal. no backtracking occurs at all when the string contains valid HTML code. The last token in the regex has been matched. The last token in the regex has been matched. Again. this is the leftmost longest match. Laziness Instead of Greediness The quick fix to this problem is to make the plus lazy instead of greedy. Now. „>” is matched successfully. Again. The requirement has been met. «>» can match the next character in the string. Now. the engine will backtrack. You can do that by putting a question markbehind the plus in the regex. the engine will backtrack. Remember that the regex engine is eager to return a match.58 Looking Inside The Regex Engine The first token in the regex is «<». Only at this point does the regex engine continue with the next token: «>». So our example becomes «<. The next token is the dot. «<. The dot matches „E”. But «>» still cannot match. the curly braces and the question mark itself. causing the engine to backtrack further. So the engine matches the dot with „E”. The minimum is one. So the match of «. You should see the problem by now. (Remember that the plus requires the dot to match only once. The next token is the dot.+» is expanded to „EM”. the backtracking will force the lazy plus to expand rather than reduce its reach. When using the negated character class. Because of greediness. The engine reports that „<EM>” has been successfully matched. The dot matches the „>”. Again. When using the lazy plus.

You will not notice the difference when doing a single search in a text editor. remember that this tutorial only talks about regex-directed engines. Text-directed engines do not backtrack. Finally. They do not get the speed penalty. or perhaps in a custom syntax coloring scheme for EditPad Pro.59 Backtracking slows down the regex engine. But you will save plenty of CPU cycles when using such a regex is used repeatedly in a tight loop in a script that you are writing. but they also do not support lazy repetition operators. .

a repetition operator. you can speed things up by using non-capturing parentheses. The colon indicates that the change we want to make is to turn off capturing the backreference. The question mark and the colon after the opening round bracket are the special syntax that you can use to tell the regex engine that this pair of brackets should not create a backreference. In the second case. You can reuse it inside the regular expression (see below). Square brackets define a character class. you can group that part of the regular expression together. e. you can optimize this regular expression into «Set(?:Value)?». This allows you to apply a regex operator. and the other letters in lowercase. EditPad Pro and PowerGREP have a unique feature that allows you to change the case of the backreference. A backreference stores the part of the string matched by the part of the regular expression inside the parentheses. round brackets also create a “backreference”. at the expense of making your regular expression slightly harder to read. Finally. How to Use Backreferences Backreferences allow you to reuse part of the regex match. The regex «Set(Value)?» matches „Set” or „SetValue”. That is. unless you use non-capturing parentheses. In the first case.g. there is no confusion between the question mark as an operator to make a token optional. slows down the regex engine because it has more work to do. \U1 inserts the first backreference in uppercase. Use Round Brackets for Grouping By placing part of a regular expression inside round brackets or parentheses. That question mark is the regex operator that makes the previous token optional.60 11. the first backreference will be empty. What you can do with it afterwards. because it did not match anything. If you searched for «EditPad (Lite|Pro)» and use “\1 version” as the replacement. and the question mark as a character to change the properties of a pair of round brackets. depends on the tool you are using. Therefore. Remembering part of the regex match in a backreference. and curly braces are used by a special repetition operator. If you do not use the backreference. the first backreference will contain „Value”. Round Brackets Create a Backreference Besides grouping part of a regular expression together. the actual replacement will be “Lite version” in case „EditPad Lite” was matched. In EditPad Pro or PowerGREP. Note that only round brackets can be used for grouping. If you do not use the backreference. you can use the backreference in the replacement text during a search-and-replace operation by typing \1 (backslash one) into the replacement text. because an opening bracket by itself is not a valid regex token. \I1 inserts it with the first letter of each word capitalized. to the entire group. \L1 in lowercase and \F1 with the first character in uppercase and the remainder in lowercase. and “Pro version” in case „EditPad Pro” was matched. or afterwards. I have already used round brackets for this purpose in previous topics throughout this tutorial. Note the question mark after the opening bracket is unrelated to the question mark at the end of the regex. This operator cannot appear after an opening round bracket. .

NET (dot net). . it will either give an error message.Value. This backreference is reused with «\1» (backslash one). you can use $1. \0 cannot be used inside a regex. In EditPad Pro or PowerGREP. to insert backreferences. it is simply empty. because that would force the engine to continuously keep an extra copy of the entire regex match. In Perl. In the replacement text. By putting the opening tag into a backreference. In . etc. the item with index zero holds the entire regex match. „bxbxb” and „cxcxc”. This object has a property called Groups. and the text in between. The . In Perl.*?</\1>». which capture the string matched by «[A-Z][A-Z0-9]» into the first backreference. To get the string matched by the third backreference in C#. A backreference cannot be used inside itself. $2. but also during the match. you can use MyMatch. or it will fail to match anything without an error message. Libraries like . Depending on your regex flavor. «([a-c])x\1x\1» will match „axaxa”. Non-capturing parentheses are not counted. This fact means that non-capturing parentheses have another benefit: you can insert them into a regular expression without changing the numbers assigned to the backreferences. etc. we can reuse the name of the tag for the closing tag. which is a collection of Group objects. to access the part of the string matched by the backreference. This regex contains only one pair of parentheses.61 Regex libraries in programming languages also provide access to the backreference. If a backreference was not used in a particular match attempt (such as in the first example where the question mark made the first backreference optional). This can be very useful when modifying a complex regular expression.NET (dot net) where backreferences are made available as an array or numbered list. you can use the magic variables $1. $2. Using an empty backreference in the regex is perfectly fine.Groups[3]. Suppose you want to match a pair of opening and closing HTML tags. the magic variable $& holds the entire regex match. Using backreference zero is more efficient than putting an extra pair of round brackets around the entire regex. etc. the second number two. you can use the Match object that is returned by the Match method of the Regex class. «([abc]\1)» will not work. only in the replacement. You can reuse the same backreference more than once. The «/» before it is simply the forward slash in the closing HTML tag that we are trying to match. scan the regular expression from left to right and count the opening round brackets. Here's how: «<([A-Z][A-Z09]*)[^>]*>. The Entire Regex Match As Backreference Zero Certain tools make the entire regex match available as backreference zero. Therefore.NET (dot net) Regex class also has a method Replace that can do a regex-based search-and-replace on a string. It will simply be replaced with nothingness. To figure out the number of a particular backreference. The first bracket starts backreference number one. Using Backreferences in The Regular Expression Backreferences can not only be used after a match has been found. you can use the entire regex match in the replacement text during a search and replace operation by typing \0 (backslash zero) into the replacement text.

«<» matches the third „<” in the string. so the engine backtracks again. But this did not happen here. The last token in the regex. This match fails. That is because in the second regex. the regex engine does not permanently substitute backreferences in the regular expression. it will read the value that was stored. A complete match has been found: „<B><I>bold italic</I></B>”. Repetition and Backreferences As I mentioned in the above inside look. «>» matches „>”. and not «B». «[^>]» does not match „>”. The first token in the regex is the literal «<». These match. The engine advances to «[A-Z0-9]» and “>”. repeated by a lazy star.62 Looking Inside The Regex Engine Let's see how the regex engine applies the above regex to the string “Testing <B><I>bold italic</I></B> text”. After storing the backreference. The next token is a dot. It will use the last match saved into the backreference each time it needs to be used. and the next token is «/» which matches “/”. These obviously match. This also means that «([abc]+)=\1» will match „cab=cab”. Every time the engine arrives at the backreference. The engine arrives again at «\1». the regex engine will initially skip this token. the previously saved match is overwritten. Again. In this case. and the dot consumes the third “<” in the string. If a new match is found by capturing parentheses. «[A-Z]» matches „B”. Note that the token the backreference. The position in the string remains at “>”. This step crosses the closing bracket of the first pair of capturing parentheses. The next token is «\1». the new value stored in the first backreference would be used. The next token is «/». «B» matches „B”. because of the star. These do not match. The next token is «[A-Z]». This fails to match at “I”. The backtracking continues until the dot has consumed „<I>bold italic”. The star is still lazy. Obvious when you look at a . this is not a problem. The engine does not substitute the backreference in the regular expression. the plus caused the pair of parentheses to repeat three times. the previous value was overwritten. „B” is stored. There is a clear difference between «([abc]+)» and «([abc])+». that's perfectly fine. „c” was stored. This means that if the engine had backtracked beyond the first pair of capturing parentheses before arriving the second time at «\1». The engine has now arrived at the second «<» in the regex. The backreference still holds „B”. because of another star. the engine proceeds with the match attempt. and the second “<” in the string. The position in the string remains at “>”. This does not match “I”. Though both successfully match „cab”. The regex engine will traverse the string until it can match at the first „<” in the string. Each time. At this point. Because of the laziness. The second time „a” and the third time „b”. the first regex will put „cab” into the first backreference. it holds «b» which fails to match “c”. so „B” it is. The regex engine also takes note that it is now inside the first pair of capturing parentheses. «<» matches „<” and «/» matches „/”. At this point. The reason is that when the engine arrives at «\1». The first time. and the engine is forced to backtrack to the dot. so the engine again backtracks. and position in the regex is advanced to «>». However. taking note that it should backtrack in case the remainder of the regex fails. The dot matches the second „<” in the string. so „b” remains. The position in the regex is advanced to «[^>]». so the engine again takes note of the available backtracking position and advances to «<» and “I”. while the second regex will only store „b”. This prompts the regex engine to store what was matched inside them into the first backreference. Backtracking continues again until the dot has consumed „<I>bold italic</I>”. and that «([abc])+=\1» will not.

So this regex will match an a followed by either «\x01» or a «b». „(” and „)”. simply type in “\1” as the replacement text and click the Replace button. Useful Example: Checking for Doubled Words When editing text. When using backreferences. . always double check that you are really capturing what you want. at least not as metacharacters. The \1 in regex like «(a)[\1b]» will be interpreted as an octal escape in most regex flavors. Parentheses and Backreferences Cannot Be Used Inside Character Classes Round brackets cannot be used inside character classes. doubled words such as “the the” easily creep in. Backreferences also cannot be used inside a character class. it is treated as a literal character. When you put a round bracket in a character class. To delete the second word. but a common cause of difficulty with regular expressions nonetheless. Using the regex «\b(\w+)\s+\1\b» in your text editor.63 simple example like this one. So the regex «[(a)b]» matches „a”. „b”. you can easily find them.

use «\k<name>» or «\k'name'». starting with one. In PHP. The numbers can then be used in backreferences to match the same text again in the regular expression. Named Capture with Python. Unfortunately. where the sharp brackets are used for HTML tags. This does not work in PHP. The second syntax is preferable in ASP code. PHP. Python and PCRE treat named capturing groups just like unnamed capturing groups.64 12.NET framework also support named capture. Python's sub() function allows you to reference a named group as “\1” or “\g<name>”. «(?P<name>group)» captures the match of «group» into the backreference “name”. The PHP preg functions offer the same functionality. RegexBuddy supports both Python's and Microsoft's style. PHP/preg. Again. and number both kinds from left to right. Named Capture with .NET style: «(?<first>group)(?'second'group)». you can use double-quoted string interpolation with the $regs parameter you passed to pcre_match(): “$regs['name']”. you can use the two syntactic variations interchangeably. or one of the . since they are based on PCRE. When doing a search-and-replace. To reference a capturing group inside the regex. Simply use a name instead of a number between the curly braces.NET offers two syntaxes to create a capturing group: one using sharp brackets. Here is an example with two capturing groups in . and offers named capture using the same syntax.NET languages.Text. or to use part of the regex match for further processing. The first syntax is preferable in strings. the Microsoft developers decided to invent their own syntax. Use Round Brackets for Grouping All modern regular expression engines support capturing groups. and the other using single quotes. no other regex flavor supports Microsoft's version of named capture. which are numbered from left to right. You can use the pointy bracket flavor and the quoted flavors interchangeably. The regex .NET's System. . Currently. PCRE and PHP Python's regex module was the first to offer a solution: named capture. and will convert one flavor of named capture into the other when generating source code snippets for Python. starting with one. Names and Numbers for Capturing Groups Here is where things get a bit ugly. you can reference the named group with the familiar dollar sign syntax: “${name}”.RegularExpressions The regular expression classes of the . By assigning a name to a capturing group. You can reference the contents of the group with the numbered backreference «\1» or the named backreference «(?P=name)». the numbering can get a little confusing. you can easily reference it by name. As you can see. where single quotes may need to be escaped. rather than follow the one pioneered by Python. The open source PCRE library has followed Python's example. In a complex regular expression with many capturing groups.

from one till four. Things are quite a bit more complicated with the . To make things simple. if you do a search-and-replace with “$1$2$3$4” as the replacement. However. If you do a search-and-replace with this regex and the replacement “\1\2\3\4”.65 «(a)(?P<x>b)(c)(?P<y>d)» matches „abcd” as expected. since the regex engine does not need to keep track of their matches. you will get “acbd”. when using . but numbers them after all the unnamed groups have been numbered. Either give a group a name. you will get “abcd”. continuing from the unnamed groups. in this case: three. So the unnamed groups «(a)» and «(c)» get numbered first. Then the named groups «(?<x>b)» and «(?<y>d)» get their numbers. The regex «(a)(?<x>b)(c)(?<y>d)» again matches „abcd”. and reference them by name exclusively. I strongly recommend that you do not mix named and unnamed capturing groups at all. Non-capturing groups are more efficient. All four groups were numbered from left to right. The . just assume that named groups do not get numbered at all. Probably not what you expected.NET framework. .NET framework does number named capturing groups from left to right.NET's regex support. or make it non-capturing as in «(?:nocapture)». from left to right. starting at one. To keep things compatible across regex flavors. Easy and logical.

matches() method in Java does not take a parameter for matching options like Pattern. while Pattern.66 13. Specifying Modes Inside The Regular Expression Sometimes. Turning Modes On and Off for Only Part of The Regular Expression Modern regex flavors allow you to apply modifiers to only part of the regular expression. the handy String. Older regex flavors usually apply the option to the entire regular expression. Regex Matching Modes All regular expression engines discussed in this tutorial support the following three matching modes: • • • /i makes the regex match case insensitive. E. . and one to turn it off. you use a modifier span. In that situation. you can add a mode modifier to the start of the regex. m/regex/i turns on case insensitivity. (?i-sm) turns on case insensitivity. The regex «(?i)te(?-i)st» should match „test” and „TEst”. You have probably noticed the resemblance between the modifier span and the non-capturing group «(?:group)». If you insert the modifier (?ism) in the middle of the regex. «(?i)ignorecase(?-i)casesensitive(?i)ignorecase» is equivalent to «(?i)ignorecase(?i:casesensitive)ignorecase». You can turn off a mode by preceding it with a minus sign. Most tools that support regular expressions have checkboxes or similar controls that you can use to turn these modes on or off.CASE_INSENSITIVE) does the same in Java. In this mode. turns off single-line mode.g. the modifier only applies to the part of the regex to the right of the modifier. the caret and dollar match before and after newlines in the subject string. The latest versions of all tools and languages discussed in this book do. Pattern. and turns on multi-line mode. (?i) turns on case insensitivity. To turn off several modes. no matter where you placed it.compile(“regex”. E. Most programming languages allow you to pass option flags when constructing the regex object. In this mode.g. E. precede each of their letters with a minus sign. /m enables "multi-line mode". the tool or language does not provide the ability to specify matching options.g. but not “teST” or “TEST”. in Perl. Not all regex flavors support this. while (?ism) turns on all three options. Technically. You can quickly test this. It is obvious that the modifier span does not create a backreference. E.g. one to turn an option on. /s enables "single-line mode". Many regex flavors have additional modes or options that have single letter equivalents. Modifier Spans Instead of using two modifiers.compile() does. but these differ widely. the non-capturing group is a modifier span that does not change any modifiers. the dot matches newlines.

9.3. and the {11} skips the first 11 fields.11.» has consumed „11. the comma does not match the “1” in the 12th field.” as well as „11. Since there is still no P.*?. But it does not give up there. Let's say the string is “1.».12. 10th.”.7.67 14.){11}P». The next token is again the dot. The dot matches the comma! However.11. A greedy quantifier will first try to repeat the token as many times as possible.8. and gradually give up matches as the engine backtracks to find an overall match.10. Finally.10. let's see why backtracking can lead to problems. again trying all possible combinations for the 9th. the P checks if the 12th field indeed starts with P. so the dot continues until the 11th iteration of «.11”.12.2.9. Continuously failing. the same story starts with the 9th iteration.”. A lazy quantifier will first repeat the token as few times as required. At that point. However.2.8. the engine backtracks to the 8th iteration.6. Greediness and laziness determine the order in which the regex engine tries the possible permutations of the regex pattern.”.3. they do not change the fact that the regex engine will backtrack to try all possible permutations of the regular expression in case no match can be found. they can change the overall regex match. this leads to a catastrophic amount of backtracking. The problem rears its ugly head when the 12th field does not start with a P. It will backtrack to the point where «^(. The customer was using the regexp «^(.12. there are more possiblities to be tried. But between each expansion.e. When the 9th iteration consumes „9.*?. this is exactly what will happen when the 12th field indeed starts with a P. expanding the match of the 10th iteration to „10.12. and 11th iterations. the 10th iteration is expanded to „10.*?. Atomic Grouping and Possessive Quantifiers When discussing the repetition operators or quantifiers.4. the regex engine can no longer match the 11th iteration of «. The lazy dot and comma match a single comma-delimited field. the regex engine will backtrack. Reaching the end of the string again.5.”. You get the idea: the possible number of combinations that the regex engine will try for each line where the 12th field does not start with a P is huge. the 10th could match just „11.){11}» had consumed „1. or even crash as the regex engine runs out of memory trying to remember all backtracking positions. and gradually expand the match as the engine backtracks through the regex to find an overall match. It does not. this regex looks like it should do the job just fine.10. subsequently expanding it to „9.*?.11.4. I explained the difference between greedy and lazy repetition. „9. Because of the double repetition (star inside {11}).”. The dot matches a comma.11.10. You can already see the root of the problem: the part of the regex (the dot) matching the contents of the field also matches the delimiter (the comma). First.11. .10.”. It backtracks to the 10th iteration.5. giving up the last match of the comma.”. Catastrophic Backtracking Recently I got a complaint from a customer that EditPad Pro hung (i. The regex engine now checks whether the 13th field starts with a P. This causes software like EditPad Pro to stop responding. Since there is no comma after the 13th field. In fact. it stopped responding) when trying to find lines in a comma-delimited text file where the 12th item on a line started with a “P”.6. At first sight.10.”. „9.7.13”. Because greediness and laziness change the order in which permutations are tried.12.

we could easily reduce the amount of backtracking to a very low level by better specifying what we wanted. Similarly. If there is no token before the group. you can use «x*+». as do recent versions of PCRE and PHP's pgreg functions. without trying further options.n}+». Using atomic grouping. «x++» is the same as «(?>x+)». The fields must not contain comma's. If backtracking is required. In the above example. Because the entire group is one token. Atomic Grouping and Possessive Quantifiers Recent regex flavors have introduced two additional solutions to this problem: atomic grouping and possessive quantifiers. To make a quantifier possessive.6. Python does not support atomic grouping.\r\n]» is not able to expand beyond the comma. Perl supports it starting with version 5.NET support atomic grouping. the engine has to backtrack to the regex token before the group (the caret in our example). and PCRE version 4 and later.68 Preventing Catastrophic Backtracking The solution is simple. and only supported by the latest versions of most regex flavors. In our example. But that is not always possible in such a straightforward manner. allowing the regex engine to fail faster. the regex must retry the entire regex at the next position in the string.0 and later. . At this time. In that case. though the JDK documentation uses the term “independent group” rather than “atomic group”. Everything between (?>) is treated as one single token by the regex engine. So the regex becomes: «^([^. make absolutely sure that there is only one way to match the same match. you should use atomic grouping to prevent the regex engine from backtracking. The latest versions of EditPad Pro and PowerGREP support both atomic grouping and possessive quantifiers.){11})P». forcing the regex engine to the previous one of the 11 iterations immediately. Note that you cannot make a lazy quantifier possessive. «x?+» and «x{m. no backtracking can take place once the regex engine has found a match for the group.*?.\r\n]*. Possessive quantifiers are a limited form of atomic grouping with a cleaner notation.4. The Java supports it starting with JDK version 1. But it will backtrack only 11 times. the engine will still backtrack.4. All versions of . If repeating the inner loop 4 times and the outer loop 7 times results in the same overall match as repeating the inner loop 6 times and the outer loop 2 times. When nesting repetition operators.2. We want to match 11 commadelimited fields. the solution is to be more exact about what we want to match. If the P cannot be found. Tool and Language Support for Atomic Grouping and Possessive Quantifiers Atomic grouping is a recent addition to the regex scene. place a plus after it. It would match the minimum number of matches and never expand the match because backtracking is not allowed. as do all versions of RegexBuddy. the above regex becomes «^(?>(. once the regex engine leaves the group. and each time the «[^. Their purpose is to prevent backtracking. you can be sure that the regex engine will try all those combinations.){11}P». possessive quantifiers are only supported by the Java JDK 1.

If you are simply doing a search in a text editor.6. the cause of this is that the token «\d» that is repeated can also match the delimiter «6». When nesting quantifiers like in the above example.69 Atomic Grouping Inside The Regex Engine Let's see how «^(?>(. you will not earn back the extra time to type in the characters for the atomic grouping.7. Now comes the difference. since possessive. then atomic grouping may make a difference.9. «\d+» will match the entire string. rather than after 30 attempts to match the caret and a huge number of attempts to try all combinations of both quantifiers in the regex. They do not speed up success. if you are smart about combined repetition. This shows again that understanding how the regex engine works on the inside will enable you to avoid many pitfalls and craft efficient regular expressions that match exactly what you want. «{11}» causes further repetition until the atomic group has matched „1.5. So far.){11})P» is applied to “1.4. so the engine backtracks to the dot. The star is not possessive.11. The most efficient regex for our problem at hand would be «^(?>((?>[^. the engine backtracks until the 6 can be matched. and declares failure.8.3. you can reduce clutter by writing «^(?>([^. Failure is declared after 30 attempts to match the caret. the amount of time wasted increases exponentially and will very quickly exhaust the capabilities of your computer.10. The engine walks through the string until the end. the engine leaves the atomic group.*?.3. only failure. If the final x in the regex cannot be matched. everything happened just like in the original. . you really should use atomic grouping and/or possessive quantifiers whenever possible. The dot matches „1”. and the comma matches too. no backtracking is allowed.2. greedy repetition of the star is faster than a backtracking lazy dot.8.\r\n]*).5. so the group's entire match is discarded and the engine backtracks further to the caret. With combined repetition.”. and just one attempt to match the atomic group. Because the group is atomic. The previous token is an atomic group. which fails. often it is not. That is what atomic grouping and possessive quantifiers are for: efficiency by disallowing backtracking. The engine now tries to match «P» to the “1” in the 12th field. In the latter case. you often can avoid the problem without atomic grouping as in the example above.12. troublesome regular expression. This fails. the increase in speed is minimal. The caret matches at the start of the string and the engine enters the atomic group. If possessive quantifiers are available. When To Use Atomic Grouping or Possessive Quantifiers Atomic grouping and possessive quantifiers speed up failure by eliminating backtracking. That is.6.2. If the regex will be used in a tight loop in an application. the regex engine backtracks once for each character matched by the star.4. «\d+6» will match „123456” in “123456789”.7. While «x[^x]*+x» and «x(?>[^x]*)x» fail faster than «x[^x]*x». «P» failed to match.){11})P». That's right: backtracking is allowed here. so the engine backtracks. With simple repetition. Sometimes this is desirable. and the match fails. Still. or process huge amounts of data. Now. The star is lazy.9. using simple repetition. Again. and is not immediately enclosed by an atomic group. But the comma does not match “1”. the amount of time wasted with pointless backtracking increases in a linear fashion to the length of the string. The engine now tries to match the caret at the next position in the string.10.\r\n]*+. so the dot is initially skipped.){11})P».11. «\d++6» will not match at all. Note that atomic grouping and possessive quantifiers can alter the outcome of the regular expression match.13”. all backtracking information is discarded and the group is now considered a single token. the regex engine did not cross the closing round bracket of the atomic group. With the former regex.

This does not match the void behind the string. If it contains capturing parentheses. (Note that this is not the case with lookbehind. They are zero-width just like the start and end of line. we have the trivial regex «u». you have to put capturing parentheses around the regex inside the lookahead. They do not consume characters in the string. Regex Engine Internals First. but only assert whether a match is possible or not. If you want to store the match of the regex inside a backreference. The difference is that lookarounds will actually match characters. with the opening bracket followed by a question mark and an equals sign. I will explain why below. Lookarounds allow you to create regular expressions that are impossible to create without them. The next character is the “u”. Collectively. When explaining character classes.70 15. the backreferences will be saved. The engine advances to the next character: “i”. and begins matching the regex inside the lookahead. Inside the lookahead. That is why they are called “assertions”. without making the u part of the match. As we already know. Negative lookahead provides the solution: «q(?!u)». However. like this: «(?=(regex))». the entire regex has matched. «q» matches „q”. The engine takes note that it is inside a lookahead construct now. So the next token is «u». These match. They are also called “zero-width assertions”. The engine notes that the regex inside the lookahead failed. The other way around will not work. it is done with the regex inside the lookahead. or that would get very longwinded without them. This causes the engine to step back in the string to “u”. All regex flavors discussed in this book support lookaround. The next token is the «u» inside the lookahead. I already explained why you cannot use a negated character class to match a “q” not followed by a “u”. . The first token in the regex is the literal «q». So it is not included in the count towards numbering the backreferences.) Any valid regular expression can be used inside the lookahead. but then give up the match and only return the result: match or no match. The negative lookahead construct is the pair of round brackets. You can use any regular expression inside the lookahead. this will cause the engine to traverse the string until the „q” in the string is matched. Note that the lookahead itself does not create a backreference. The next token is the lookahead. which supports lookahead but not lookbehind. and „q” is returned as the match. and start and end of word anchors that I already explained. Let's try applying the same regex to “quit”. The positive lookahead construct is a pair of round brackets. these are called “lookaround”. At this point. and discards the regex match. «q(?=u)» matches a q that is followed by a u. Because the lookahead is negative. let's see how the engine applies «q(?!u)» to the string “Iraq”. with the opening bracket followed by a question mark and an explanation point. The engine notes success. Lookahead and Lookbehind Zero-Width Assertions Perl 5 introduced two very powerful constructs: “lookahead” and “lookbehind”. Positive lookahead works just the same. The position in the string is now the void behind the string. because the lookahead will already have discarded the regex match by the time the backreference is to be saved. Positive and Negative Lookahead Negative lookahead is indispensable if you want to match something not followed by something else. this means that the lookahead has successfully matched at the current position. The exception is JavaScript.

and finds out that the “m” does not match «a». with the opening bracket followed by a question mark. The engine again steps back one character. It finds a “t”.) Again. the “h”. the current position in the string remains at the “m”. and the engine starts again at the next character. you could use «\b\w+(?<!s)\b». The next character is the first “b” in the string. More Regex Engine Internals Let's apply «(?<=a)b» to “thingamabob”. which cannot match here. It will not match “cab”. In this case. Important Notes About Lookbehind The good news is that you can use lookbehind anywhere in the regex. but will match the „b” (and only the „b”) in “bed” or “debt”. not only at the start. the engine has to start again at the beginning. using an exclamation point instead of an equals sign. The engine steps back and finds out that „a” satisfies the lookbehind. To lookahead was successful. Since there are no other permutations of this regex. So this match attempt fails. Again. Since «q» cannot match anywhere else. “less than” symbol and an equals sign. and put a token after it. using negative lookbehind. «b» matches „b”. The engine cannot step back one character because there are no characters before the “t”. The next token is «b». so the positive lookbehind fails again. the engine temporarily steps back one character to check if an “a” can be found there. Again. All remaining attempts will fail as well. The next character is the second “a” in the string. The construct for positive lookbehind is «(?<=text)»: a pair of round brackets. If you want to find a word not ending with an “s”. «(?<=a)b» (positive lookbehind) matches the „b” (and only the „b”) in „cab”. but works backwards. But «i» cannot match “u”. «(?<!a)b» matches a “b” that is not preceded by an “a”. the lookbehind tells the engine to step back one character. It tells the regex engine to temporarily step backwards in the string. Negative lookbehind is written as «(?<!text)». So the lookbehind fails. The lookbehind continues to fail until the regex reaches the “m” in the string. the engine reports failure. «q» matches „q” and «u» matches „u”. Positive and Negative Lookbehind Lookbehind has the same effect. The engine steps back. I have made the lookahead positive. to make sure you understand the implications of the lookahead. the successful match inside it causes the lookahead to fail. the match from the lookahead must be discarded. Let's take one more look inside. and see if an “a” can be matched there.71 Because the lookahead is negative. Let's apply «q(?=u)i» to “quit”. (Note that a negative lookbehind would have succeeded here. This is definitely not the same as . The engine starts with the lookbehind and the first character in the string. because there are no more q's in the string. and notices that the „a” can be matched there. to check if the text inside the lookbehind can be matched there. but does not match “bed” or “debt”. It matches one character: the first „b” in the string. The positive lookbehind matches. so the engine continues with «i». Because it is zero-width. and the entire regex has been matched successfully. so the engine steps back from “i” in the string to “u”.

including infinite repetition. The latter will also not match single-letter words like “a” or “I”. You can use alternation.4. plus finite repetition. The reason is that regular expressions do not work backwards.72 «\b\w+[^s]\b». . The last regex. the semantics of applying a regular expression backwards are currently not well-defined.NET framework can apply regular expressions backwards. I will leave it up to you to figure out why. though. alternation and character classes inside lookbehind. These regex flavors recognize the fact that finite repetition can be rewritten as an alternation of strings with different. EditPad Pro and PowerGREP.0 of the . RegexBuddy. has a double negation (the \W in the negated character class). but fixed lengths. and will allow you to use any regex. The only regex flavor that I know of that currently supports this is Sun's regex package in the JDK 1. This means you can still not use the star or plus. I recommend you use only fixed-length strings. Until that happens. Therefore. Microsoft has promised to resolve this in version 2. plus alternation with strings of different lengths. (Hint: «\b» matches between the apostrophe and the “s”). Some regex flavors support the above. JavaScript does not support lookbehind at all. including those used by Perl 5 and Python. The correct regex without using lookbehind is «\b\w*[^s\W]\b» (star instead of plus. many regex flavors. so only literals and character classes can be used. The bad news is that you cannot use just any regex inside a lookbehind. which works correctly. Not to regex engines. The string must be traversed from left to right. the .NET framework. and \W in the character class). PHP. This includes PCRE. but only if all options in the alternation have the same length. Technically. Double negations tend to be confusing to humans. You can use any regex of which the length of the match can be predetermined. inside lookbehind. Even with these limitations. But each string in the alternation must still be of fixed length. You cannot use repetition or optional items. I find the lookbehind easier to understand. This means you can use literal text and character classes. the former will match „John” and the latter „John'” (including the apostrophe). only allow fixed-length strings. lookbehind is a valuable addition to the regular expression syntax. Finally. However. When applied to “John's”. Personally. Finally. Therefore. but you can use the question mark and the curly braces with the max parameter specified. some more advanced flavors support the above. the regular expression engine needs to be able to figure out how many steps to step back before checking the lookbehind.

At this position will the regex engine attempt the remainder of the regex. The engine will then backtrack. We just specify all the options and hump them together using alternation: «cat\w{3}|\wcat\w{2}|\w{2}cat\w|\w{3}cat». But this method gets unwieldy if you want to find any word between 6 and 12 letters long containing either “cat”. Matching a 6-letter word is easy with «\b\w{6}\b». Combining the two. the last «\b» in the regex is guaranteed to match where the second «\b» inside the lookahead matched. This is at the second letter in the 6-letter word we just found. If not. To make this clear. in the 6letter word. Actually. is a very powerful concept. If «cat» cannot be matched. . Second. reducing the number of characters matched by «\w*». where the lookahead will fail. we get: «(?=\b\w{6}\b)\b\w*cat\w*\b». First. matches only when the current character position in the string is at the start of a 6-letter word in the string. “dog” or “mouse”. until «cat» can be matched. at the next character position in the string. a bit more practical example. or a lookbehind is preceded by another piece of regex. The confusing part is that the lookaround is zero-width. and therefore the lookahead. Unfortunately. After that. I would like to give you another. we want a word that is 6 letters long. Our double-requirement-regex has matched successfully. the engine has no other choice but to restart at the beginning of the regex. If «cat» can be successfully matched. then the regex will traverse part of the string twice. So when the regex inside the lookahead has found the 6-letter word. we know that «\b» matches and that the first «\w*» will match 6 times. the second «\w*» will consume the remaining letters. Matching a word containing “cat” is equally easy: «\b\w*cat\w*\b». Let's say we want to find a word that is six letters long and contains the three subsequent letters “cat”. So if you have a regex in which a lookahead is followed by another piece of regex. This sub-regex. Easy enough. and the engine will continue trying the regex from the start at the next character position in the string. causing the engine to advance character by character until the next 6-letter word. At each character position in the string where the regex is attempted. because lookaround is a bit confusing. if any. we can match this without lookaround. the word we found must contain the word “cat”. it is often underused by people new to regular expressions.73 16. Lookaround to The Rescue In this example. Because we already know that a 6-letter word can be matched at the current position. the current position in the string is still at the beginning of the 6-letter word. Testing The Same Part of The String for More Than One Requirement Lookaround. The lookahead is zero-width. which I introduced in detail in the previous topic. we basically have two requirements for a successful match. Easy! Here's how this works. the lookahead will fail. the engine will first attempt the regex inside the positive lookahead.

but if a 6-letter word does not contain “cat”. and therefore does not change the result returned by the regex engine. leaving: «(?=\b\w{6}\b)\w*cat\w*». it will match 6 letters and then backtrack. up to and including “cat”. So we have «(?=\b\w{6}\b)\w{0. the resulting match would be the start of a 6-letter word containing “cat”.3}». “dog” or “mouse” into the first backreference. But we know that in a successful match.9}(cat|dog|mouse)\w*». optimization involves the first «\b». as I did above. If we omitted the «\w*». Since it is zero-width. what would you use to find any word between 6 and 12 letters long containing either “cat”.74 Optimizing Our Solution While the above regex works just fine. So the final regex is: «\b(?=\w{6}\b)\w{0.12}\b)\w{0. . Very easy. A More Complex Problem So. so it does not contribute to the match returned by the regex engine. This regex will also put “cat”. This is not a problem if you are just doing a search in a text editor. But we can optimize the first «\w*». Note that making the asterisk lazy would not have optimized this sufficiently. there can never be more than 3 letters before “cat”. Since it is zero-width itself. You can discover these optimizations by yourself if you carefully examine the regex and follow how the regex engine applies it. minor.3}cat\w*». As it stands. it would still cause the regex engine to try matching “cat” at the last two letters. One last. there's no need to put it inside the lookahead. The lazy asterisk would find a successful match sooner. and even at one character beyond the 6-letter word. Remember that the lookahead discards its match.3}cat\w*». once you get the hang of it. But optimizing things is a good idea if this regex will be used repeatedly and/or on large chunks of data in an application you are developing. So we can optimize this to «\w{0. instead of the entire word. we can remove them. I said the third and last «\b» are guaranteed to match. it is not the most optimal solution. “dog” or “mouse”? Again we have two requirements. which we can easily combine using a lookahead: « \b(?=\w{6. we cannot remove it because it adds characters to the regex match. Though the last «\w*» is also guaranteed to match. at the last single letter.

Substitute «wanted». we can do without lookahead. and «stop» as the regex matching the end of the section. we found a match before a section rather than inside a section. The final regular expression will be in the form of «wanted(?=insidesection)». I will use «wanted» as a substitute for the regular expression that we are trying to match inside the section. and end with the section stop. this will not work. Note that these two rules will only yield success if the string or file searched through is properly translated into sections. The reason is that this regular expression consumes the entire section. Second. and «start» as the regex matching the start of the section. «start» and «stop» with the regexes of your choice. Lookbehind must be of fixed length.*?». we must be able to match «stop» after matching «wanted». How do we know if we matched «wanted» inside a section? First. You may be tempted to use a combination of lookbehind and lookahead like in «(?<=start. Because of the negative lookahead inside the star. When we apply the regex again to the same string or file. The star is obviously not of fixed length. So we need a way to match „wanted” without matching the rest of the section.*?stop)». So inside the lookahead we need to look for a series of unspecified characters that do not match the start of a section anywhere in the series. we repeat zero or more times with the star. If “wanted” occurs only once inside the section.*?)wanted(?=. it will continue after „stop”. each match of «start» must be followed exactly by one match of «stop». The entire section is included in the regex match. The final regular expression becomes: «wanted(?=((?!start). That is. you can easily build a regex to do a search and replace on HTML files.75 17.*?stop» would do the trick.*?wanted. To keep things simple. If not. This. the star will also stop at the start of a section. Example: Search and Replace within Header Tags Using the above generic regular expression. A title tag starts with «<H[1-6]» and . If we could. Since we do not know in advance how many characters there will be between “start” and “wanted”. we need to use «. Finding Matches Only Inside a Section of The String Lookahead allows you to create regular expression patterns that are impossible to create without it. First we match the string we want. However. at which point stop cannot be matched and thus the regex will fail. we found a match after a section rather than inside a section. «start. In a regex. However. The dot and negative lookahead match any character that is not the first character of the start of a section. but only inside title tags. we must not be able to match «start» between matching «wanted» and matching «stop». One example is matching a particular regex only inside specific sections of the string or file searched through. we need to match the end of the section.)*?stop». not after „wanted”. Effectively. The regex engine will refuse to compile this regular expression. After this. this will not work if “wanted” occurs more than once inside a single section. replacing a certain word with another. So we have to resort to using lookahead only. This is possible with lookahead. and then we test if it is inside the proper section. the lazy star will continue to repeat until the end of the section is reached.)*?stop)». because lookahead is zero-width. this is written as: «((?!start). I used a lazy star to make the regex more efficient.

76 ends with «</H[1-6]>». I omitted the closing > in the start tag to allow for attributes.)*?</H[1-6]>)». But lookahead is what we need here. or negative lookbehind. Escaping the < takes care of the problem. I did that because some regex flavors interpret «(?!<» as identical to «(?<!». . You may have noticed that I escaped the < of the opening tag in the final regex. So the regex becomes «wanted(?=((?!\<H[1-6]).

this makes a lot of sense in the context of a text editor. Continuing at The End of The Previous Match The anchor «\G» matches at the position where the previous match ended. the position where the last match ended is a “magical” value that is remembered separately for each string variable. All in all. This means that you can use «\G» to make a regex continue in a subject string where another regex left off. so the match fails. specify the continuation modifier /c. rather than the end of the previous match. Applying it again matches „e”. where «\G» matches at the position of the text cursor. Applying «\G\w» to the string “test string” matches „t”. But that position is not followed by a word character. This way you can parse the tags in the file in the order they appear in the file.. you could parse an HTML file in the following fashion: while ($string =~ m/</g) { if ($string =~ m/\GB>/c) { } elsif ($string =~ m/\GI>/c) { } else { } } # Bold # Italics # .. the only place in the string where «\G» matches is after the second t. If a match attempt fails.. without having to write a single big regex that matches all tags you are interested in. During the fifth attempt. \G Magic with Perl In Perl. E. The regex in the while loop searches for the tag's opening bracket. rather than at the end of the previous match result. During the first match attempt. The 3rd attempt yields „s” and the 4th attempt matches the second „t” in the string. EditPad Pro will select the match. All this is very useful to make several regular expressions work together. To avoid this. This is the case with EditPad Pro.etc. End of The Previous Match vs Start of The Match Attempt With some regex flavors or tools.77 18. The fifth attempt fails. the stored position for «\G» is reset to the start of the string. «\G» matches at the start of the match attempt.. When a match is found. The result is that «\G» matches at the end of the previous match result only when you do not move the text cursor between two searches. and the regexes inside the loop check which tag we found. «\G» matches at the start of the string in the way «\A» does. and move the text cursor to the end of the match. The position is not associated with any regular expression.g. .

The Matcher is strictly associated with a single regular expression and a single subject string. the position for «\G» is remembered by the Matcher object.g. «\G» will then match at this position. in Java. . What you can do though is to add a line of code to make the match attempt of the second Matcher start where the match of the first Matcher ended. E.78 \G in Other Programming Langauges This flexibility is not available with most other programming languages.

then the regex engine will attempt to match the then part. If you use a lookahead as the if part.79 19. the syntax becomes «(?(?=regex)then|else)». If-Then-Else Conditionals in Regular Expressions A special construct «(?ifthen|else)» allows you to create conditional regular expressions. you can use the lookahead and lookbehind constructs. the else part is attempted instead. The syntax consists of a pair of round brackets. If the if part evaluates to true. The opening bracket must be followed by a question mark. the if and then parts are clearly separated. This part can be followed by a vertical bar and the else part. there is no need to use parentheses around the then and else parts. then the regex engine will attempt to match the then or else part (depending on the outcome of the lookahead) at the same position where the if was attempted. like in «(?(?=condition)(then1|then2|then3)|(else1|else2|else3))». Otherwise. Remember that the lookaround constructs do not consume any characters. For the then and else. Using positive lookahead. . and the vertical bar with it. If you want to use alternation. For the if part. Because the lookahead has its own parentheses. you can use any regular expression. immediately followed by the then part. immediately followed by the if part. You may omit the else part. you will have to group the then or else together using parentheses. Otherwise.

g. . Now it is instantly obvious that this regex matches a date in yyyy-mm-dd format./. I guess you will agree that regular expressions can quickly become rather cryptic. That makes the comments really stand out. Therefore. EditPad Pro and PowerGREP can apply syntax coloring to regular expressions while you write them. The regex engine ignores everything after the «(?#» until the first closing round bracket.](?#month)(0[1-9]|1[012])[.80 20. many modern regex flavors allow you to insert comments into regexes. Adding Comments to Regular Expressions If you have worked through the entire tutorial. Some software. as long as it does not contain a closing round bracket. enabling the right comment in the right spot to make a complex regular expression much easier to understand.](?#day)(0[1-9]|[12][0-9]|3[01])». I could clarify the regex to match a valid date by writing it as «(?#year)(19|20)\d\d[/. E. The syntax is «(?#comment)» where “comment” is be whatever you want. such as RegexBuddy.

Sign up to vote on this title
UsefulNot useful